Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    fixed, uniform, discrete

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    aboration
    0.74
     experimentation
    0.74
    caution
    0.70
     dépasse
    0.69
     Zudem
    0.69
     Zweifel
    0.68
    agues
    0.67
     ನೋಡ
    0.67
     supplémentaire
    0.67
    वर्स
    0.66
    POSITIVE LOGITS
     uniform
    1.08
     fixed
    1.00
    Uniform
    0.93
     uniformly
    0.91
     preset
    0.90
    uniform
    0.88
     pre
    0.86
     predetermined
    0.85
    fixed
    0.81
    Fixed
    0.81
    Act Density 0.375%

    No Known Activations